ANSYS Solutions for the Built Environment - Present Capability and Future Trends

Phil Stopford

Phil.Stopford@ansys.com

Presentation to MAGIC Partners Meeting

8th Dec 2020

• DECEMBER 2019

Software &Systems

Platform

Thermal management Multiphase flow Fluid-structure External Aerodynamics Internal Flow Electronics Cooling Mixing Combustion

Vibration Optimization Strength Analysis Thermal Analysis Durability Impact FatigueReliability Composites

Antennas – 5G Microwave Electronics Cooling Electric Motors Power Electronics Radar Cross Section Radio Frequency Signal Integrity Power Integrity

SOC PowerIntegrity IP Power Reliability IC AnalysisTRL Power Efficiency SOC ReliabilityCPS System CO-Design Substrate Noise Automotive IC 7NM Chip Design

System prototype Electric Drives Power ElectronicsAutonomous Vehicles ADASBattery ManagementSystem Functional Safety Analysis Embedded Control Software

Additive manufacturing Guided workflows Instant simulation Direct Modeling Concept Design

Optics

Bright Light Appearance Material Evaluation Human Ergonomics Dynamic DrivingSoundAnalysis Perceived Quality

2

Ansys offers a wide simulation platform

Fluids

FLUIDS

Structures

STRUCTURES

Electromagnetics

ELECTROMAGNETICS OPTICALMISSION-CRITICAL

EMBEDDED SOFTWARE

SEMICONDUCTORPOWER

3D DESIGN

PLATFORMS

Leading Construction Companies Use ANSYS

Wind Engineering

Building Stability & Integrity

HVACConstructionEquipment

Cement Manufacturing

Old Paradigms

Design and analysis are correlation and rule based

Single physics in engineering silos

Few Design Points Studied

Hardware and parallel computation limitations

Many specialized niche simulation tools

Product development is test dominant

Certification through test

ANSYS Vision For Simulation In The Construction Industry

Courtesy BDP

New Paradigms

Single-Platform Multiphysics Simulation Including fluids,structure, seismic, electromagnetic

Collaborative Multi-user and Multiscale Simulation (component to subsystem and system)

High-Performance Computing including cloud and supporting Remote Infrastructure

Higher fidelity simulation

Failure analysis

Multi disciplinary design optimization

Consistence early adoption of simulation in building design

Certification through analysis

CFD for Built Environment

5

Simulation Methods

• High Quality Meshing Algorithms‐ Unstructured

‐ Structured Hex

‐ Advanced, Automatic Sizing Controls

• Volume

• Surface

‐ High Quality Prism Layers

• Accurate Physics and Numerics− Segregated and Coupled CFD Solvers

• Fast transient and steady solutions

− High order discretisation schemes

− Best-in-class turbulence modelling capabilities

• Automatic wall treatment inc. roughness

• RANS, SAS, DES, LES, SBES

− Extensive physics options such as

• Aeroacoustics

• FEA coupling (Fluid-Structure Interaction)

• Aerothermal

• Solar loading

• Combustion/Species/humidity modeling

• Particles and multiphase

– Tsunami/flood modelling with VoF

– Sand build-up

• Electromagnetics Coupling

Octree Hexcore Polyhedral

Fast and Scalable CFD & FEA Calculations

• CFD scalable from <10K cells per partition to beyond

1bn cell cases

• Automatic load balancing

• CPU and GPU solver acceleration used

Serial – 1 Core

Parallel – 32 Cores

32 X Speedup

Built Environment Application Areas

• Internal/External Aerodynamics‐ Pedestrian comfort

• Perform sweeps of wind direction

• Steady flow or gusting

‐ Pollution dispersion

• Inert or reacting

‐ Wind loading

‐ Aeroacoustics

‐ Plume Effects

‐ Heating and Ventilation

Plume visualisation from different wind directions

Built Environment Application Areas

• Health and Safety− Fire suppression

− Smoke Management

− HVAC

− Structural vibration and thermal response

©British Crown Copyright 2007/MOD.

Published with the permission of the Controller of Her Britannic

Majesty's Stationery Office

Shedding from chimney with and without helical strakes applied Thermal beam deformation due to fire

• Extended Physics− Solar Loading

− Structural deformations from wind loading

− Blast wave propagation

− Wave modelling

− Sand Build-up

Built Environment Application Areas

Damage

Blast Wave Propagation

Wind Loading

Total Deformation

Sand Coloured by HeightSolar Loading in an Atrium

Case Study: September 11th, World Trade Centre

‐ Investigating the engineering consequences of the attacks

Courtesy of the National Oceanic and Atmospheric Administration, and the U.S. Environmental Protection Agency, National Exposure Research Laboratory, and the U.S. EPA Environmental Modeling and Visualization Laboratory, Lockheed-Martin Operations Support

Plume Development: September 11th, World Trade Centre

http://gallery.ensight.com/Categories/Fluent/11118289_qLxkGT/1/701169171_JoAmZ#701169171_JoAmZ

Grey = smokeGreen = particles

Recent Developments

13

• Scale-Adaptive Simulation (SAS)– Reverts to RANS when mesh too coarse

• Detached Eddy Simulation (DES)– Similar behaviour to SAS

• Large Eddy Simulation (LES)– Smagorinsky-Lilly and Dynamic model

– WALE model

– Algebraic Wall Modeled LES (WMLES)

– Dynamic kinetic energy subgrid model

• Embedded or Zonal LES (ELES)– LES in fixed zone with turbulence generator upstream

– Not suitable for globally unstable flows, e.g. bluff bodies

• Stress-Blended Eddy Simulation (SBES)– RANS-LES blended at level of Reynolds stress

– Faster switch to LES mode

Scale-Resolving Turbulence Models

GEKO RANS Model: Introducing Free Coefficients

( ) ( )

+

+−=

+

jk

t

j

k

j

j

x

k

xkCP

x

kU

t

k

( ) ( )

+

+

+−=

+

j

t

j

jj

k

j

j

xx

xx

kFFCP

kFC

x

U

t

23

2

2211

( ),

,max Real

tCS

k

=

• CSEP – changes separation behaviour

• CMIX – changes spreading rates of free shear flows

• CNW – changes near-wall behaviour

• CJET – Optimizes free jet flows

• CCORNER – Affects corner flows

• CCURVE – Curvature correction

The functions F1, F2, and F3 contain 6 free coefficients that can be changed without affecting basic validation:

Coefficients can be changed to give best fit to experiment

𝑢𝑖′𝑢𝑗

′ → 𝑢𝑖′𝑢𝑗

′ −𝐶𝐶𝑜𝑟𝑛𝑒𝑟1.2 𝑡

𝑀𝐴𝑋 0.3𝜔, (𝑆2 + 2)/2𝑆𝑖𝑘𝑘𝑗 −𝑖𝑘𝑆𝑘𝑗

• GPU acceleration can give LES results in real time on large-scale models

• Demo simulation for a stadium showing instantaneous vorticity

• Uses a single GeForce GTX 1080 with 8GB GPU and 2560 CUDA Cores

• Not sufficiently accurate for detailed analysis but can be useful for first design

Discovery Live: Real-Time LES Simulation using GPU

Reduced Order Model (ROM) as a simplification of a high-fidelity dynamical model

Benefits of ROM

Reduced Order Model (ROM)

Reduced simulation time (think 10-100x)• Ideal for Design of Experiments (DoE)/

Parameter sweep• Integration in Twin Builder for system simulation• Runtime generation for near real-time applications

Reduced storage size• Reduce the required storage size dramatically

Reuse 3D model• Utilize validated 3D physics in system model• Help increase the 3D solver footprint

Techniques for range of physics including Fluid Flow, Thermal, Mechanical and EM

DX-ROM , Static ROM Builder and Twin Builder Dynamic ROM are few ANSYS ROM technologies

Fluent CFD Simulation:3 hours on 12 cores

ROM SimulationRealtime

Summary

• Ansys development driven by customer requirements

• Adoption of LES for Built Environment has been slow over last decade− RANS still seen as best option in many cases

− LES ~100x more expensive and does not always give better results than RANS

− LES primarily used for visualisation, aeroacoustics and particle dispersion

• Ansys has focused on developments in− Geometry and meshing: More complexity and larger meshes

− Parallel computation and cloud computing: > 1 billion cells

− Turbulence modelling: GEKO RANS, and hybrid RANS-LES (SBES)

− Reduced order modelling (Twin Builder)